MULTILINGUAL

MULTILINGUAL

AI-based language-learning mobile app

Hypothetical Solo Project

Summary

Multilingual, a hypothetical language-learning app that incorporates AI to tailor its educational content to each user specifically. Each exercise learns from the user’s behaviour in order to cater the next exercise to their ability. Multilingual offers an AI chatbot that the user can have conversations with, wherein the bot learns from these conversations and can create exercises for the user to do based on a topic of conversation that was discussed. Although nothing beats full immersion, not everybody can afford to travel to learn a language - but almost everybody can access a mobile app.

Problem

Traditional language-learning apps lack personalisation, making it challenging for users to learn effectively based on their unique needs and preferences. Existing platforms offer standardised lessons that don't adapt to individual learning styles and struggles. This one-size-fits-all approach often leads to frustration and limited progress for learners.

Goal

Design a language-learning app that utilises AI to tailor its educational content to the user’s specific learning needs.


Find out if AI is the key to solving some of the user pain points uncovered in this study.

Process

  • Research
  • Interviews
  • Competitive Audit
  • Ideation
  • Wireframing
  • Prototyping
  • UI design
  • Testing


Role

Researcher

UX Designer

UX Writer

Research

Learning objectives

Trying to find out the needs and frustrations of people who currently experience language-learning apps available on the market.

Methodology

I researched Facebook groups and conducted a survey to establish proper user pain points to influence my design choices. My survey was aimed at 100 people of any age and gender who had to qualify in the following conditions:


1. Have an interest in learning a language.

2. Are currently using apps such as Duolingo, Babbel, Rosetta Stone, etc to learn.


Below are two examples of multiple choice questions asked in my 8-question survey.


High level findings

  • Finding opportunities for practice - 70%
  • Developing listening and comprehension skills - 60%
  • Grammar rules - 50%
  • Pronunciation difficulties - 40%
  • Retaining and recalling vocabulary effectively - 40%
  • Understanding and using proper word order in sentences - 30%
  • Other (answers included handwriting practice for languages such as Japanese, and learning idioms) - 20%
  • Vocabulary building -10%


High level findings

  • Real-life dialogues or conversations - 100%
  • Videos/audio recordings with transcripts - 90%
  • Authentic texts, articles, or news stories - 80%
  • Interactive exercises or quizzes - 60%
  • Gamified elements or challenges - 60%
  • Cultural insights and multimedia content - 40%

Pain points

  • People struggle to find time to incorporate learning a language in their day-to-day lives.


  • People find that a lot of language-learning tools are repetitive and lack genuine reflection of real-world usage.


  • Mastering vocabulary retention and recall can be challenging, especially when most exercises readily provide the answer, such as multiple-choice options in language-learning apps.
Simple Decorative Divider Elements | Minimalist

Competitive audit

Below you can see my insights about three Multilingual competitors - both direct and indirect. Researching and evaluating other businesses was key to understanding how to make Multilingual stand out as its own tool in a saturated language-learning app market. The competitive audit allowed me to figure out what features should/shouldn’t be incorporated in Multilingual.

Duolingo

Competitor type

Direct online competitor

Business size

Valued at over USD$2.4 billion

Price

$ - offers a freemium model

Unique Value Proposition

Gamified and interactive language learning experience with a large user community.

Target audience

Broad and encompasses language learners of all ages and proficiency levels. It is particularly popular among casual learners, beginners, and individuals seeking to learn a new language in a fun and gamified manner.

Product offering

Duolingo offers gamified language courses with interactive lessons, exercises, and a community feature, catering to learners of various languages.

SpanishDict

Competitor type

Indirect online competitor

Business size

Recognised platform with substantial user base and international presence.

Price

$ - offers a freemium model

Unique Value Proposition

Comprehensive Spanish language resources including a dictionary, grammar guides, and community interaction.


Target audience

Individuals who are specifically interested in learning the Spanish language. It caters to learners of all levels, from beginners to advanced, providing comprehensive resources to support Spanish language acquisition.


Product offering

SpanishDict provides comprehensive Spanish learning resources including a dictionary, grammar guides, audio pronunciations, and a community forum for learners to engage with native speakers and ask questions.


Rosetta Stone

Competitor type

Direct online competitor

Business size

Valued at over USD$1 billion.


Price

$$$ - higher-end pricing


Unique Value Proposition

Immersive language learning with an emphasis on real-life context and pronunciation.


Target audience

Individuals who prefer a more immersive and natural language learning approach. Often used by individuals seeking to develop a high level of proficiency in a particular language, including serious language learners, students, and professionals.


Product offering

Offers immersive language learning courses that utilise visual and audio techniques, providing interactive lessons, immediate feedback, and a clear progression path for learners.


Key takeaways

  • All apps encourage the user to set a ‘learning goal’, where the user would select an amount of time a day they would like to be learning a language. The app also prompted them to turn on notifications to make sure the user was reminded to reach their daily goal.


  • I found it inconvenient that Rosetta Stone's lessons demanded a landscape screen orientation, whereas the rest of the app was designed for portrait orientation. The exercises were also very repetitive - one of the sentences being taught came up 18 times in one lesson.


  • I liked that all of the apps’ lessons very interactive and all of the lessons incorporated reading, writing, listening and speaking exercises.
Simple Decorative Divider Elements | Minimalist

Personas, user journey map & user flow

Paper and digital wireframes

Low fidelity prototype

Using my digital wireframes, I created the first Lo-Fi prototype. I also conducted a moderated usability study with 3 participants and asked them questions, established their goals, KPI’s, methodology and their characteristics. These newly gained insights helped me improve my design, which I utilised in my high fidelity mockups. In hindsight, I should have re-done the low fidelity wireframes and conducted a second usability study before going straight in to the high fidelity mockups. This is something I will be sure to do with my future designs, so as to not skip any important steps and incur a sunk cost fallacy.

You can view the Lo-Fi prototype here

Usability study findings

Multiple lesson types

  • Finding: Users express a strong desire for a variety of lesson types and formats within the app.


  • Discussion: Participants noted that while the app's standard lessons (e.g., vocabulary, grammar) were valuable, they craved diversity in learning experiences. Some users preferred interactive exercises, while others preferred listening and speaking activities. Users with varying learning styles and goals wanted options to tailor their learning journey.


  • Improvement Opportunity: Introduce additional lesson types such as conversational practice, cultural insights, and interactive games to cater to different learning preferences. Implement a "Lesson Library" where users can choose from a variety of lesson types to keep their language learning experience engaging and well-rounded. Allow users to set preferences for the types of lessons they want to receive as part of their daily learning plan.




Word of the day feature

  • Finding: Users appreciate and benefit of a "Word of the Day" feature, which presents them with a new word in their chosen language every day.


  • Discussion: Participants reported that a "Word of the Day" feature would not only help expand their vocabulary but also provided a daily dose of motivation and engagement. They think it particularly useful in reinforcing their learning and keeping language learning top of mind.


  • Improvement Opportunity: Enhance the "Word of the Day" feature by incorporating audio pronunciation and example sentences to provide more context and improve retention. Additionally, allow users to review past words of the day in a dedicated section for continued reinforcement.


UI Kit

Improving the design

While working on Multilingual’s Hi-Fidelity prototype, I started designing more polished wireframes with imagery and colour. After the usability tests, I decided to make some changes in the homepage’s design changing some colours to pass the WCAG standards, adding a ‘Word of the day’ section, incorporated a ‘course outline’ instead of a ‘completed course’ section to encourage users to keep learning and return to the app. I also removed the AI Chatbot from the homepage as it was already accessible directly from the bottom navigation.

Before

After

Course outline section

Word of the day section instead of AI chatbot access

Final design

High fidelity mockups

Key learnings

  • User-Centered Design is essential: Understanding the needs and preferences of my target users is paramount. Prioritising their experience over personal preferences or assumptions leads to better outcomes.


  • Research matters: Conducting thorough user research helped me uncover insights and make informed design decisions. It's reiterated to me that it’s the foundation of user-centered design.


  • Wireframing and prototyping is crucial: Creating prototypes allowed me to visualise and test my ideas early, saving time and resources in the long run. It helped me identify issues and refine my design before development.


  • Feedback is valuable: Constructive criticism can lead to improvements and a better end product.


  • Consistency is key: Maintaining consistency in design elements like colours, fonts, and layout enhances the user experience and makes my product look and feel polished.


  • Accessibility is vital: Ensuring my designs are accessible to all users, including those with disabilities as this not only improves inclusivity but is often required by law.


  • Passion and empathy: UX design is about more than just creating aesthetically pleasing interfaces. It's about understanding and empathising with users' needs and frustrations.

Next steps

  • Usability testing and feedback incorporation: Schedule more usability testing sessions to gather feedback on the current design to identify pain points and areas for improvement.


  • Iterate further on the design: Continue iterating on the design based on user feedback and insights gained from testing and make necessary adjustments to enhance the user experience.


  • User Flow and Information Architecture: Review and optimise the user flows and information architecture of the project to ensure that users can easily navigate the system and find the information they need.

Conclusion

Does the use of AI really improve someone’s ability to learn a language effectively?

In summary, this hypothetical case study has allowed me to explore the potential of integrating AI into a mobile language-learning app. Through user research and prototyping, I've identified key design principles for a user-centered experience. While this project may not lead to a real product, the insights gained are valuable for future design endeavors in the field of technology and education, and have taught me that users appreciate content that is tailor-made to their own individual needs. AI integration in a mobile language-learning app can aid in solving some user pain points, especially in terms of ensuring that the exercises aren’t overly repetitive and disengaging the user while they’re trying to learn. While AI may not assist users in finding time for language learning in their daily lives, a fun and engaging app can motivate them to return and engage with the learning experience consistently.

THANK YOU!

Want to get in touch?

Pigeon

Pigeons are outdated, here are some better ways to reach me: